1. Field of the Invention
The present invention relates to an optical division multiplexing transmission and reception method and an optical division multiplexing transmission and reception device and, more particularly, to a method and device that make it possible for wavelength division multiplexing (WDM) transmission and reception and optical code division multiplexing (OCDM) transmission and reception to coexist. Further, the present invention relates to a method, and device that make it possible for optical time division multiplexing (OTDM) transmission and reception and OCDM transmission and reception to coexist.
2. Description of Related Art
In order to increase the speed or capacity of optical fiber communications, an optical division multiplexing technology that transmits a plurality of channels' worth of optical pulse signals all together on one optical fiber transmission line has been investigated. As means for the optical division multiplexing, WDM, which separates channels by means of the wavelengths of the optical pulses constituting the optical pulse signals, OTDM, which separates channels by means of the time slots that are occupied by the optical pulses that constitute the optical pulse signals, and OCDM, which separates channels by means of pattern matching of encoded optical pulse signals have each been researched.
Therefore, WDM and OTDM will be described first with reference to FIG. 1. FIG. 1 is a schematic block constitutional view of an optical division multiplexing transmission and reception device in which a transmission section 100 and reception section 200 are constituted linked by an optical fiber 105, which is a transmission line. The number of channels is denoted as n.
The transmission section 100 comprises a transmitter 101, transmitter 102, and transmitter 103 for the first to nth channels respectively. Further, a multiplexing device 104 that mixes and multiplexes the optical pulse signals of the first to nth channels outputted from the transmitters is provided. The transmitter 101, transmitter 102, and transmitter 103 convert electrical signal 110, electrical signal 111, and electrical signal 112 of the first to nth channels into the optical pulse signal 120, optical pulse signal 121, and optical pulse signal 122 respectively. The optical pulse signals outputted from the respective transmitters of the first to nth channels are mixed and multiplexed by the multiplexing device 104 and outputted as an optical division multiplexing signal 126.
The optical division multiplexing signal 126 outputted from the multiplexing device 104 is transmitted to the reception section 200 through propagation via the optical fiber 105 constituting the transmission line.
The reception section 200 comprises a receiver 107, receiver 108, and receiver 109 for the first to nth channels respectively. Further, a separator 106 that separates the optical division multiplexing signal 126 that is inputted to the receivers is provided.
Here, the optical division multiplexing transmission and reception device shown in FIG. 1 will first be described as a WDM transmission and reception device. In the case of the WDM transmission and reception device, light of a different wavelength for each channel is allocated as the carrier wave for the respective channel information. That is, in the case of a WDM transmission and reception method and a device that implements this method, the wavelength of the carrier-wave light plays the role of an identifier for identifying the channel.
An optical coupler, for example, is used as the multiplexing device 104. Further, an optical element rendered by combining an optical coupler and optical wavelength filter, for example, or an optical element that has a wavelength separation function such as an Array Waveguide Grating (AWG) is used as the separator 106. Therefore, the inputted multiple-wavelength optical division multiplexing signal 126 is separated by the separator 106 into wavelengths that are allocated to each channel which are then outputted. As a result, optical pulse signals of wavelengths that are allocated to the respective channels are supplied to the receiver 107, receiver 108, and receiver 109.
OTDM will be described next. The optical division multiplexing transmission and reception device shown in FIG. 1 is described as an OTDM transmission and reception device. Optical pulse signals that are modulated to the RZ (Return to Zero) format are outputted from the transmitter 101, transmitter 102, and transmitter 103. When the optical pulse signals of the first to nth channels are mixed by the multiplexing device 104, adjustment of the timing for inputting the optical pulses to the time slots provided for all the channels is performed by using a variable delay line or the like, for example.
A combination of an optical coupler that intensity-divides the optical division multiplexing signal 126 outputted from the multiplexing device 104 according to the number of channels, and an optical modulator that allows light of only a specified time slot to be transmitted, for example, is used for the separator 106. The separator 106 separates the optical division multiplexing signal 126 into each channel and the optical pulse signal 123 of the first channel, the optical pulse signal 124 of the second channel, and the optical pulse signal 125 of the nth channel are supplied to the receiver 107, receiver 108, and receiver 109 of the respective channels. The receiver 107, receiver 108, and receiver 109 convert the O/E converted optical pulse signals into electrical pulse signals and receive the electrical pulse signals 113, 114, and 115 of the respective channels.
The dispositional relationship of the optical pulses that constitute the respective optical pulse signals on the wavelength axis and time axis of WDM and OTDM respectively will now be described with reference to FIGS. 2A and 2B. FIG. 2A shows an aspect in which the respective channels are arranged divided on the wavelength axis for WDM. Further, FIG. 2B shows an aspect in which the respective channels are allocated to each positional slot divided into time slots on the time axis for OTDM.
When FIGS. 2A and 2B are referenced, it can be seen that the wavelength bandwidth is used in the WDM case and time slots designated through division on the time axis are used in the OTDM case in order to allocate the respective channels. That is, WDM and OTDM are systems in which one of the physical resources such as wavelengths or time slots in which one channel is divided on the wavelength axis or time axis is used occupied.
It can be seen from the above description that, for OTDM, the wavelength of the light source is basically not a problem. However, in order to increase the number of multiplexed channels in OTDM, the time slots allocated to the respective channels must be shortened and the half-value width on the time axis of the optical pulses constituting the optical pulse signals must also be narrowed.
On the other hand, in the case of WDM, separation from multiplexed optical pulse signals into optical pulse signals for each channel can be implemented by a passive light component with a wavelength separation function. Further, an optical pulse signal in the RZ format or an NRZ (Non-Return to Zero) format signal can be applied as an optical pulse signal and the number of multiplexed channels can be changed even without changing the transmission speed (bit rate). In addition, merits such as the fact that asynchronous multiplexing can also be implemented are combined. As a result, WDM-related research has been vigorously performed until now and is currently put to practical use.
Recently, research into OCDM, which is a method different from WDM and OTDM mentioned above, has begun as an optical multiplexing method. The merit of OCDM is that there is no need to occupy one of the physical resources such as wavelengths or time slots in which one channel is divided on the wavelength axis or time axis as per WDM and OTDM.
The constitution and functions of an example of an OCDM device will be described to with reference to FIGS. 3A to 3E (See N. Wada, et al., “Error-free transmission of 2-channel×2.5 Gbit/s time-spread/wavelength-hop OCDM using fibre Bragg grating with supercontinuum light source”, ECOC'99, September 1999) and Japanese Patent Application Laid Open No. 2000-209186, for example). The OCDM device shown in FIG. 3A is a constitution in which a transmission section 300 and a reception section 400 are linked by a transmission line 310. The transmission line 310 is an optical fiber. In order to avoid a complicated description, FIG. 3A shows a device that assumes transmission and reception on two channels. It is clear from the following description that an OCDM device that permits transmission and reception on three or more channels can be similarly implemented by increasing the number of channels.
The transmission section 300 comprises a first-channel encoder 303, a second-channel encoder 304, and a multiplexer 307. The first channel encoder 303 encodes a first-channel optical pulse signal 301 by means of code supplied by Code 1 and outputs the result as the first-channel encoded optical pulse signal 305. The second-channel encoder 304 similarly encodes a second-channel optical pulse signal 302 by means of code supplied by Code 2 and outputs the result as the second-channel encoded optical pulse signal 306.
FIG. 3B shows the time waveforms of the first- and second-channel optical pulse signals. The optical pulses constituting the first- and second-channel optical pulse signals contain light components of different wavelengths λ1, λ2, and λ3. In order to illustrate this, rectangles that surround the numbers 1, 2, and 3 that identify the wavelengths λ1, λ2, and λ3 are expediently shown stacked on the same time. Here, optical pulse signals that are constituted by optical pulses containing wavelengths of three different types are assumed and illustrated. However, the types of wavelengths that are generally contained in optical pulses are not limited to three types and the following description is similarly established in cases where two or more than two types are established.
The fact that the optical pulses contain light components of the different wavelengths λ1, λ2, and λ3 and so forth means that, when the optical pulses are arranged broken down on the wavelength axis, that is divided, division is into optical pulses whose center wavelengths are λi, λ2, and λ3 and so forth. Further, optical pulses that comprise a single optical wavelength component that is obtained by wavelength breakdown of optical pulses constituted comprising a plurality of light components will also be referred to subsequently as a chip pulse.
Hereinafter, it is assumed that optical pulses containing different wavelength components are shown by stacking rectangles surrounding identification numbers denoting the wavelengths of the wavelength components on the same time. Further, in order to identify the first-channel optical pulses and second-channel optical pulses, the second-channel optical pulses is shown shaded.
FIG. 3C shows a first-channel encoded optical pulse signal 305 and second-channel encoded optical pulse signal 306 with respect to the time axis. As shown in FIG. 3C, when the first-channel encoded optical pulse signal 305, for example, is considered, the optical pulses constituting the first-channel optical pulse signal 301 are divided by the encoder 303 into optical pulses (chip pulses) with the center wavelengths λ1, λ2, and λ3 and arranged after undergoing time spreading on the time axis. The same is true of the second-channel encoded optical pulse signal 306. However, because the code (Code1) established for the first-channel encoder and the code (Code2) established for the second-channel encoder are different codes, the positions for arranging the respective chip pulses arranged on the time axis of the first- and second-channel encoded optical pulse signals are different.
Thus, the encoding performed by the device shown in FIG. 3A is a method that performs encoding by subjecting the optical pulses to time spreading on the time axis and then division into optical pulses (chip pulses) with the center wavelengths λ1, λ2, and λ3 that constitute the optical pulses and is therefore known as encoding by means of time-spreading/wavelength-hopping code. That is, encoding of the first- and second-channel input optical pulse signals 301 and 302 by means of time-spreading/wavelength-hopping code by means of the first-channel encoder 303 and second-channel encoder 304 is performed.
FIG. 3D shows an optical code division multiplexing signal 308 rendered by multiplexing the first-channel encoded optical pulse signal 305 and second-channel encoded optical pulse signal 306 by means of the multiplexer 307. The multiplexer 307 affords a multiplexer function of multiplexing optical signals of a plurality of channels. A chip pulse array that constitutes the first-channel encoded optical pulse signal 305 and a chip pulse array that constitutes the second-channel encoded optical pulse signal 306 are stacked on the same time axis as shown in FIG. 3C. Here, in order to be able to identify the chip pulses constituting the first-channel encoded optical pulse signal and the chip pulses constituting the second-channel encoded optical pulse signal, the latter second-channel encoded optical pulse signal is shaded.
The optical code division multiplexing signal 308 is sent to the reception section 400 through propagation via the transmission line 310. The reception section 400 comprises a splitter 410, and a first-channel decoder 413 and second-channel decoder 414. The splitter 410 subjects the optical code division multiplexing signal 308 to intensity division, supplying one of the split signals to the first-channel decoder 413 as a split optical code division multiplexing signal 411 and the other to the second-channel decoder 414 as a split optical code division multiplexing signal 412.
The first-channel decoder 413 plays back the split optical code division multiplexing signal 411 by decoding same by means of code that is supplied by Code 1 and outputs the decoded signal as a first-channel optical pulse signal 415. The second-channel decoder 414 similarly plays back the split optical code division multiplexing signal 412 by decoding same by means of code that is supplied by Code 2 and outputs the decoded signal as a second-channel optical pulse signal 416. The optical pulse signals that are played back by the respective decoders are also subsequently called the decoded optical pulse signals.
FIG. 3E shows that the optical code division multiplexing signal 308 undergoes intensity division for each of the first and second channels by means of the splitter 410 with which the reception section 400 is provided and shows the decoded optical pulse signals that are decoded by the first-channel decoder 413 and second-channel decoder 414 for the first and second channels.
First, the first-channel decoded optical pulse signal 415 will be described. In an aspect that represents the optical intensity with respect to the time axis of the first channel in FIG. 3E, chip pulses that originate in the second-channel optical pulse signal are shown by shaded rectangles that surround the numbers identifying the wavelength and chip pulses that originate in the first-channel optical pulse signal do not have shaded rectangles that surround the numbers identifying the wavelengths.
The chip pulses originating in the first-channel optical pulse signal are chip pulses that are generated encoded by code supplied by Code 1 and, therefore, if the chip pulses are decoded by means of code supplied by the same Code 1, the respective chip pulses are arranged to occupy the same positions on the time axis with the time delays provided during encoding exactly offset. That is, the original optical pulse signal is played back as an autocorrelation waveform.
Looking at the diagram representing the optical intensity with respect to the time axis of the first channel in FIG. 3E, unshaded rectangles that surround the numbers 1, 2, and 3 are stacked on the same time. On the other hand, shaded rectangles that surround the numbers 1, 2, and 3 appear as a mutual interlayer waveforms that are arranged dispersed in different positions on the time axis. Shaded rectangles that surround the numbers 1, 2, and 3 are chip pulses that originate in the second channel and are chip pulses that constitute encoded optical pulse signals that are encoded by means of Code 2. That is, because the encoded optical pulse signal component comprising chip pulses originating in the second channel is executed by means of codes which are different for encoding and decoding, the time lag provided during encoding is not offset during decoding and is constituted as a mutual interlayer waveform that is time-dispersed once again.
In a drawing that represents the optical intensity with respect to the time axis of the second channel in FIG. 3E, a relationship results that is the inverse of that described above. That is, chip pulses originating in the second channel form autocorrelation waveforms and chip pulses originating in the first channel form mutual correlation waveforms. This is because the constitution is such that the second channel is encoded by code that is provided by means of Code 2 and decoded by code that is provided by Code 2.
Because the intensity-divided optical code division multiplexing signal 412 is decoded by means of code that is provided by Code 2, the time lag provided during encoding of the chip pulses originating in the first channel that are encoded by means of code provided by Code 1 contained in the optical code division multiplexing signal 412 is not offset during decoding and the chip pulses are constituted once again as time-spread mutual correlation waveforms. On the other hand, the time lag provided during encoding of the chip pulses originating in the second channel that are encoded by code that is provided by Code 2 contained in the optical code division multiplexing signal 412 is offset during decoding and constituted as an autocorrelation waveform.
As described hereinabove, the decoded optical pulse signal 415 of the first channel and the decoded optical pulse signal 416 of the second channel are established as the sum of autocorrelation waveforms and mutual correlation waveforms respectively. As shown in FIG. 3E, because the peak intensity is different in the autocorrelation waveform and the mutual correlation waveform (the peak of the autocorrelation waveform is larger), if the mutual correlation waveform component is removed by subjecting the peak values of the waveforms to a threshold value judgment in which the size of the peak values are judged with respect to a preset threshold value, only the autocorrelation waveform component is removed. If the respective autocorrelation waveform components of each channel can be extracted, the autocorrelation waveforms are the respective optical pulse signals that are played back and, therefore, if the optical pulse signals are converted to electrical signals, the transmitted information can be received.
Encoding and decoding methods include a method for encoding an optical pulse signal that uses light of a single wavelength in addition to the time-spreading/wavelength-hopping method. With this method, encoding is performed by arranging the optical pulses that constitute the optical pulse signal on a time axis by means of breakdown into chip pulses with a phase difference provided between the respective chip pulses (See P. C. Teh, et al. “Demonstration of a Four-Channel WDM/OCDMA System Using 255-Chip 320-Gchip/s Quarternary Phase Coding Gratings” IEEE, Photonics Technology Letters., vol. 14, No. 2, pp. 227-229, February 2002), for example). This encoding is also known as time-spreading encoding.
A Super Structure Fiber Bragg Grating (SSFBG) is known as an example of means for implementing the encoding and decoding. The structure and operation of an FBG optical encoder will now be described with reference to FIGS. 4A and 4B. In FIG. 4A, an aspect in which the refractive index distribution structure and the refractive index variation of the core of the optical fiber in which an SSFBG is formed is shown divided into an upper view and a lower view. As shown in the upper view of FIG. 4A, the inputted optical pulse is inputted to the SSFBG from the left side of FIG. 4A and the chip pulse array thus generated is also outputted from the left side. In the case of the SSFBG shown in FIG. 4A, because units FBG G1, FBG G2, and FBG G3 are arranged in series, code of code length 3 is established for the SSFBG. Hereinafter, an SSFBG that is constituted with a plurality of units FBG arranged in series will also be known simply as an FBG.
The refractive index modulation cycles (also called the ‘grating pitch)’ of the units FBG G1, FBG G2, and FBG G3 are Λ1, Λ2, and Λ3 respectively as shown by the lower view of FIG. 4A. Generally, there is the relation λ=2nΛ between the refractive index modulation cycle A and the Bragg reflection wavelength λ. Here, n is the average refractive index of FBG. That is, the Bragg reflection wavelength λ of unit FBG is determined by establishing the grating pitch Λ of unit FBG.
Here, when a plurality of units FBG with different grating pitches are arranged in series in one optical fiber, light (also known as ‘Bragg reflection light’ hereinbelow) of wavelengths corresponding with the grating pitch is obtained from each unit FBG. The Bragg reflection light that is reflected from the respective units FBG is reflected with different time lags in accordance with the points at which the units FBG are disposed. The use of this fact is encoding using FBG time spreading waveform hopping.
A constitutional example of an FBG optical encoder will now be described with reference to FIG. 4B. The optical encoder shown in FIG. 4B is constituted comprising an FBG 352 and an optical circulator 350. The encoded optical pulses are inputted from the input port 348 on the left-hand side of FIG. 4B to the FBG 352 via the optical circulator 350 as input light. Because the FBG 352 comprises units FBG G1, FBG G2, and FBG G3, Bragg reflection light of different wavelengths reflected from the respective units FBG is reflected. The Bragg reflection light is outputted as encoded optical pulses from an input port 354 on the right-hand side of FIG. 4B via the optical circulator 350.
Means constituted by combining an AWG (Array Waveguide Grating) and optical delay line are also known in addition to the abovementioned FBG as an optical encoder that is capable of implementing time-spreading wavelength hopping encoding (See S. Yegnanarayanan, et al., “An incoherent wavelength hopping/time spreading code-division multiple access system”, ECOC'99, September 1999), for example).
Procedures for removing the autocorrelation waveform component by separating the autocorrelation waveform component and the mutual correlation waveform component from the optical pulse signal decoded on the reception side include a time gate method in addition to the abovementioned method that uses a threshold value judgment. A time gate method is a method that uses time gate means that transmit signals only in a time zone in which the autocorrelation waveform passes by performing time adjustment so that the mutual correlation waveform overlaps the autocorrelation waveform.
As time gate means, a time gate method that uses an electro-absorption modulator (EA modulator) is known (See Naoki Minato et al., IEICE OCS2003-24, pages 49 to 54, May 2003, for example). That is, the transmission rate of the EA modulator increases by the time zone through which the autocorrelation waveform passes and, as a result of slight control in the time zone through which the mutual correlation waveform component passes, a time gate is implemented. The control of the transmission rate of the EA modulator employs a clock signal.
Further, as time gate means, a time gate method that uses an SOA (Semiconductor Optical Amplifier) is known (See K. Kitayama et al., “Optical Code Division Multiplexing (OCDM) and Its Applications to Photonics Networks”, IEICE Trans. Fundamentals, vol. E82-A, No. 12 pp. 2616-2626, December 1999), for example). This method first extracts an optical clock from a portion of the signals that are decoded by using a mode synchronization semiconductor laser. Thereafter, the decoded signals and optical clock are inputted to the SOA and the SOA produces the four-wave mixing effect in sync with the optical clock. Further, the time gate means are implemented such that only optical pulses overlapping the time zone in which the SOA is in the ON state can be transmitted by the SOA as a result of the four-wave mixing effect that is produced in sync with the optical clock.
As mentioned hereinabove, the OCDM has the characteristic that there is not necessarily a need for one channel to occupy one of the physical resources (wavelength bandwidth and time slot or the like). On the other hand, with WDM, there is a need to allocate a different wavelength bandwidth to each channel. Further, with OTDM, it is necessary to allocate a different time slot to each channel on the time axis.
Furthermore, with OCDM, the code that was used during encoding must be known in order to decode the encoded optical pulse signal that was sent encoded on the reception side. Hence, unless the code used to encode the transmitted optical pulse signal is published, a third party who is unaware of the code is unable to decode the encoded optical pulse signal. This is because optical communications using OCDM are highly stable in comparison with optical communications using WDM or OTDM or the like.
In addition, merits of OCDM include the fact that an increase in the number of channels can be dealt with flexibly. For example, with WDM, in order to increase the number of channels within the restricted communication wavelength bandwidth, the wavelength bandwidth allocated to all the channels must be reset by narrowing the wavelength bandwidth allocated to each channel. Further, similarly with OTDM, in order to increase the number of channels within the restricted communication wavelength bandwidth, the time slots allocated to all the channels must be reset by narrowing the width of the time slots allocated to each channel. Before optical communications are to be performed by means of WDM or OTDM, the light source, wavelength separator, and so forth constituting the optical communication device that is used must be changed.
On the other hand, in the case of OCDM, if the size of the ratio between the peak value of the mutual correlation waveform and the peak value of the autocorrelation waveform can be secured to the extent of being able to extract an autocorrelation waveform by removing the mutual correlation waveform from the decoded optical pulse signal, channels can be added simply by adding code types. That is, the addition of new channels can be implemented simply by adding an encoding section and decoding section for which new codes corresponding with the newly added channels have been set without changing the constituent parts of channels other than the added channels of the optical communication device.
If a method and device that allow WDM transmission and reception and OCDM transmission and reception to be implemented at the same time without changing the hardware resources of the WDM optical multiplexing communication system can be implemented, the number of channels that can be received can be increased. Alternatively, if a method and device that allow OTDM transmission and reception and OCDM transmission and reception to be implemented at the same time can be implemented, the number of channels that can be transmitted and received can be increased. In addition, OCDM transmission and reception afford the above-mentioned benefits.